首页> 外文期刊>International Journal of Innovative Computing Information and Control >REDUCTION OF PROCESSING TIMES FOR TEMPORAL SUBTRACTION ON LUNG CT IMAGE EMPLOYING OCTREE ALGORITHMS
【24h】

REDUCTION OF PROCESSING TIMES FOR TEMPORAL SUBTRACTION ON LUNG CT IMAGE EMPLOYING OCTREE ALGORITHMS

机译:减少肺CT图像使用八位算法的时间减法处理时间

获取原文
获取原文并翻译 | 示例
           

摘要

The temporal subtraction image, which can be obtained by subtracting previous image from current one, is useful for visual screening in clinical field. The temporal subtraction technique removes normal structures, e.g., blood vessel. Hence, it can enhance interval changes such as the new lesions and the changes of existing abnormalities on medical images. Recently, several temporal subtraction methods have been proposed for thoracic medical images. In temporal subtraction, image registration technique is required for correcting displacement between current image and previous one. However, efficient image registration technique of temporal subtraction for MDCT (Multi Detector-row CT) has not been proposed because of the complication of deformation in 3 dimensional region. In this paper, we propose a new efficient computer aided diagnosis (CA D) algorithms for detection of lung nodules which are obtained by temporal subtraction for thoracic MDCT images. We have tried to reduce the computational time for the temporal subtraction image by use of octree algorithms on 3-dimensional image space. To evaluate our method, we have applied the method to 4 MDCT dataset and confirmed its efficiency.
机译:可以通过从当前图像中减去先前图像获得的时间相减图像可用于临床领域的视觉筛查。时间减法技术去除了正常结构,例如血管。因此,它可以增强间隔变化,例如新病变和医学图像上现有异常的变化。近来,已经提出了几种用于胸部医学图像的时间相减方法。在时间减法中,需要使用图像配准技术来校正当前图像和上一张图像之间的位移。然而,由于3维区域中的变形的复杂性,尚未提出用于MDCT(多探测器行CT)的时间相减的有效图像配准技术。在本文中,我们提出了一种新的有效的计算机辅助诊断(CA D)算法,用于通过时间相减获得的胸部MDCT图像来检测肺结节。我们已经尝试通过在3维图像空间上使用八叉树算法来减少时间相减图像的计算时间。为了评估我们的方法,我们将该方法应用于4个MDCT数据集并确认了其效率。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号